J ournal of Antimicrobial Chemotherapy (1998) 42, 591595

Introduction
Many essential oils have been advocated for use in com-
plementary medicine for bacterial and fungal infections
including boils, acne, gingivitis and vaginal candidiasis.
14
However, few of the many claims of therapeutic efcacy
have been validated adequately by either in-vitro testing
or in-vivo clinical trials. Unless these claims have been
substantiated scientically, complementary medicines are
unlikely to secure a place in conventional healthcare. One
essential oil for which some data exist is tea tree oil. This
oil is steam-distilled from leaves of the Australian native
plant Melaleuca alternifolia and has been used widely as a
topical antiseptic in Australia for almost 80 years.
5
Recent
investigations in Australia have conrmed that tea tree
oil has in-vitro activity against a wide range of Gram-
positive and Gram-negative bacteria, including Propioni -
bacterium acnes and methicillin-resistant Staphylococcus
aureus.
68
The overgrowth of Candidaspp. in the vagina is termed
vaginal candidiasis, or more commonly, thrush.
9
The
normal vaginal ora is dominated by Lactobacillus spp.
which acidify the environment and prevent pathogenic
organisms from colonizing the vagina and causing disease.
1 0
Many factors, including antibiotics, oral contraceptives,
infection and dietary changes, can alter vaginal ora,
providing opportunity for the overgrowth of Candida
s p p .
4 , 1 0 , 1 1
Approximately 8090% of thrush cases are caused
by Candida albicans with Candida glabrata, Candida
tropicalis, Candida parapsilosis and others being respon-
sible for the remainder.
11
Essential oils have been recommended for use as home
remedies for the treatment of vaginal candidiasis by
numerous books and articles in the popular press.
4,12
Products containing essential oils, designed for the treat-
ment of vaginal conditions and formulated specically for
intra-vaginal use, are available in the UK, USA, Australia
and Europe.
13,14
As yet, no data have been published as to
the in-vitro or in-vivo efcacy of such products containing
tea tree oil.
The present investigation examines the in-vitro
susceptibility of Candida spp. to a range of essential oils,
including tea tree oil and intra-vaginal tea tree oil
products.
591
In-vitro activity of essential oils, in particular Melaleuca alternifolia
(tea tree) oil and tea tree oil products, against Candidaspp.
Katherine A. Hammer
a
, Christine F. Carson
a
and Thomas V. Riley
a,b
a
Department of Microbiology, The University of Western Australia, The Queen Elizabeth I I Medical Centre,
Nedlands, Western Australia, 6009;
b
The Western Australian Centre for Pathology and Medical Research,
The Queen Elizabeth I I Medical Centre, Nedlands, Western Australia, 6009
The in-vitro activity of a range of essential oils, including tea tree oil, against the yeast
candida was examined. Of the 24 essential oils tested by the agar dilution method against
Candida albicans ATCC 10231, three did not inhibit C. albicans at the highest concentration
tested, which was 2.0% (v/v) oil. Sandalwood oil had the lowest MIC, inhibiting C. albicans at
0.06%. Melaleuca alternifolia (tea tree) oil was investigated for activity against 81 C. albicans
isolates and 33 non-albicans Candida isolates. By the broth microdilution method, the
minimum concentration of oil inhibiting 90% of isolates for both C. albicans and non-albicans
Candida species was 0.25% (v/v). The minimum concentration of oil killing 90% of isolates was
0.25% for C. albicans and 0.5% for non-albicans Candida species. Fifty-seven Candida isolates
were tested for sensitivity to tea tree oil by the agar dilution method; the minimum concen-
tration of oil inhibiting 90% of isolates was 0.5%. Tests on three intra-vaginal tea tree oil
products showed these products to have MICs and minimum fungicidal concentrations
comparable to those of non-formulated tea tree oil, indicating that the tea tree oil contained in
these products has retained its anticandidal activity. These data indicate that some essential
oils are active against Candida spp., suggesting that they may be useful in the topical
treatment of supercial candida infections.
1998 The British Society for Antimicrobial Chemotherapy
J AC
K. A. Hammer et al.
Materials and methods
Organisms
One hundred and fourteen C a n d i d a spp. isolates were
o b t a i n e d from the Department of Microbiology at The
University of Western Australia, and the I nfection Control
Unit and Bacteriology and Mycology Section of The
Western Australian Centre for Pathology and Medical
Research. Reference isolates consisted of C. albicans
ATCC 10231, ATCC 90028, ATCC 90029, C. parapsilosis
ATCC 90018, S. aureus NCTC 6571 and Escherichia coli
NCTC 10418. The clinical isolates comprised C. albicans
(n 78), C. glabrata (n 12), Candida guilliermondii
(n 1), C. parapsilosis(n 13), Candida pseudotropicalis
(n 1), C. tropicalis (n 4), and Candida stellatoidea
(n 1). Clinical isolates were identied by the germ tube
test
15
and an I D32C strip (bioMrieux Vitek, I nc.,
Hazelwood, MO, USA) for non-albicansisolates.
Essential oils and intra-vaginal tea tree oil products
All essential oils except tea tree oil were kindly provided
by Sunspirit Oils Pty. Ltd, Byron Bay, NSW, Australia.
Tea tree oil (batch 93/04) was kindly supplied by Aus-
tralian Plantations Pty. Ltd, Wyrallah, NSW, Australia.
The tea tree oil complied with the I nternational Standard
I SO 4730 and the concentrations of terpinen-4-ol and
1,8-cineole, as determined by gas chromatographymass
spectrometry, were 37.1% and 3.2%, respectively.
I ntra-vaginal tea tree oil products were provided by
Australian Bodycare Corporation Pty. Ltd, Currumbin,
QLD, Australia and Phytopharmaceutical Products Pty.
Ltd, Sydney, NSW, Australia.
I noculum preparation
Each isolate was inoculated into MuellerHinton broth
(MHB) (Unipath Ltd, Basingstoke, UK) which was then
incubated overnight at 35C with shaking. Overnight
cultures were diluted to give nal inocula for agar dilutions
of approximately 10
3
and 10
4
cfu/spot, for Candida spp.
and bacteria, respectively. The nal inoculum for broth
microdilution and product evaluation tests was approx-
imately 5 10
5
cfu/mL for all organisms. Viable counts
were performed to conrm the inoculum size for broth and
product tests.
Agar dilution method
Agar plates were prepared by adding 0.5% (v/v) Tween 20
(Sigma Chemical Co., St Louis, MO, USA) to molten
MuellerHinton agar (MHA) (Unipath), and then adding
the essential oil in doubling dilutions ranging from 2% to
0.03% (v/v). Control plates of MHA with Tween 20 but no
oil and a blood agar (Unipath) plate were included. Plates
were dried for 30 min and then inoculated with each
isolate in duplicate using a multipoint replicator (Mast
Laboratories Ltd, Liverpool, UK). Plates were incubated
aerobically at 35C, immediately after inoculating. MI Cs
were determined at 24 h for bacteria and 48 h for Candida
spp. The MI C was dened as the lowest concentration of
oil inhibiting visible growth. The presence of one or two
colonies was disregarded. The minimum concentration of
oil which inhibited at least 90% of the isolates tested was
dened as the MI C
90
.
C. albicans ATCC 10231 was used for tests against the
24 essential oils. S. aureusand E. coli were used as control
strains when testing Candida spp. isolates against tea tree
oil by the agar dilution method.
Broth microdilution method
A range of doubling dilutions of tea tree oil from 2% to
0.03% (v/v) was prepared in MHB in a 96-well microtitre
tray. Tween 80 (Sigma) was included at a nal concen-
tration of 0.001% (v/v) to enhance oil solubility. After
inoculation and incubation at 35C for 48 h, subcultures of
10 L were taken from each well and spot inoculated on to
MHA. Subcultures were incubated aerobically and MI Cs
and minimum fungicidal concentrations (MFCs) deter-
mined. The MI C was dened as the lowest concentration
of oil resulting in the maintenance or reduction of the
inoculum, and the MFC as the lowest concentration of oil
resulting in the death of 99.9% of the inoculum. The MI C
9 0
was determined as described in the agar dilution method,
while the MFC
9 0
was dened as the concentration of tea
tree oil fungicidal for at least 90% of the isolates tested.
I solates were tested in duplicate and were retested if
resultant MI C or MFC values differed. C. albicans A T C C
10231 was used in all broth microdilution tests as a control.
I n-vitro testing of tea tree oil products
Tea tree oil products were diluted in sterile distilled water
so that after inoculation, the nal concentrations of tea
tree oil ranged from 2.0% to 0.03% (w/v). Each dilution
was inoculated with the test organism, C. albicans ATCC
10231, in double strength MHB to a nal concentration of
approximately 5 10
5
cfu/mL. After incubation at 35C
for 24 h, 100 L samples were taken, diluted one in ten
and spread on to blood agar. These subcultures were
incubated at 35C for 2448 h, then MI Cs and MFCs were
determined as described in the section above on the Broth
microdilution method.
Results
The MI Cs obtained for the 24 essential oils tested are
shown in Table I . No oil inhibited C. albicansATCC 10231
at the lowest concentration tested, which was 0.03%. All
592
Activity of essential oils against Candidaspp. in vitro
except cedarwood, sweet almond and evening primrose
inhibited C. albicans ATCC 10231 at concentrations of
2.0%.
MI C and MFC data for tea tree oil, obtained by the agar
and broth dilution methods are given in Table I I . Using
agar dilution, the MI C
9 0
was 0.5%, for all C a n d i d as p e c i e s
and strains tested, while MI Cs for the control isolates
S. aureus and E. coli were 0.5% and 0.25%, respectively.
Using broth microdilution, both the MI C
90
and MFC
90
for C. albicans were 0.25%. The MI C
90
and MFC
90
for
C. parapsilosis and C. glabrata were 0.25% and 0.5%,
respectively. The MI C and MFC of tea tree oil against the
control isolate C. albicansATCC 10231 were both 0.25%.
Results for the testing of tea tree oil products are shown
in Table I I I . All products showed MI Cs and MFCs similar
to non-formulated tea tree oil.
Discussion
There are data available referring to the anticandidal
activity of the more familiar essential oils such as pepper-
mint and thyme.
1619
Some oils, such as frankincense
593
Table I. MI Cs of 24 essential oils tested against
C. albicansATCC 10231 using the agar
dilution method
Common name Botanical name MI C (%v/v)
Sandalwood Santalum album 0.06
Lemongrass Cymbopogon citratus 0.12
Spearmint Mentha spicata 0.12
Oregano Origanum vulgare 0.12
Bay Pimenta racemosa 0.12
Clove Syzigium aromaticum 0.12
Petitgrain Citrus aurantium 0.25
Coriander Coriandrum sativum 0.25
Citronella Cymbopogon nardus 0.25
Tasmanian lavender Lavandula angustifolia 0.25
French lavender Lavandula angustifolia 0.5
Peppermint Mentha piperita 0.5
Basil Ocimum basilicum 0.5
Sage Salvia ofcinalis 0.5
Celery Apium graveolens 1.0
Frankincense Boswellia carteri 1.0
Ylang ylang Cananga odorata 1.0
Bergamot Citrus bergamia 1.0
Eucalyptus Eucalyptus fruticetorum 1.0
Lemon Citrus limon 2.0
Juniper J uniperus communis 2.0
Cedarwood Cedrus atlantica 2.0
Evening primrose Oenothera biennis 2.0
Sweet almond Prunus dulcis 2.0
Table II. Cumulative MI Cs and MFCs for Candidaspp. tested by agar dilution (n 57)
and broth microdilution (n 114) methods
Tea tree oil concentration (%v/v)
Organism Method Parameter 0.12 0.25 0.5
C. albicans agar MI C 0 21
a
24
broth MI C 48 81 81
broth MFC 2 81 81
C. glabrata agar MI C 1 12 12
broth MI C 4 11 12
broth MFC 1 4 12
C. parapsilosis agar MI C 0 13 14
broth MI C 9 14 14
broth MFC 0 11 14
Other Candidaspp.
b
agar MI C 1 5 7
broth MI C 5 7 7
broth MFC 2 7 7
a
MI C
90
and MFC
90
values are in bold type.
b
C. guilliermondii (n 1), C. pseudotropicalis(n 1), C. tropicalis(n 4), C. stellatoidea(n 1).
Table III. MI Cs and MFCs of tea tree oil products
against C. albicansATCC 10231
Product (% tea tree oil) MI C (%w/v) MFC (%w/v)
Gel (3%) 0.25 0.5
Gel (10%) 0.12 0.25
Pessary (10%) 0.25 0.5
K. A. Hammer et al.
and evening primrose, appear not to have been tested
previously against Candida spp. Comparison of our data
with anticandidal data reported previously showed most
results to be similar.
1619
The data obtained in this study
show that essential oils such as sage and sandalwood,
which have been suggested for use in antiseptic or anti-
microbial capacities,
12,16
possess anticandidal properties in
vitro.
There are difculties with comparing published results
of the antimicrobial activity of some essential oils. These
arise where the common name but not the botanical name
is specied and where no data are given about the
chemical composition of the oil. Also, there are many
different methods used to investigate antimicrobial
activity and the results obtained by these methods are not
always directly comparable. These reasons may, in part,
explain the differences in results obtained by different
research groups.
Tea tree oil has activity against a wide range of Gram-
positive and Gram-negative bacteria.
8
Less information is
available about the spectrum of activity and mode of
action against Candida spp. Previous reports have
indicated MI Cs of 0.04%,
16
0.2%
20
and 0.44%
21
for C.
albicans, and 0.22% for Candida spp.
21
Our data support
the above-mentioned reports and indicate that Candida
spp. are susceptible to concentrations of 0.5% tea tree
oil. I n addition, there was little inter-species variation in
susceptibility and all Candida spp. tested were uniformly
susceptible.
Several products utilizing tea tree oil as an active
ingredient have been developed for the treatment of
vaginal candidiasis. These include intra-vaginal pessaries,
gels and douche preparations. Products such as these have
been available for many years
13,14,22
and offer alternatives
to the more conventional therapies of azole derivatives,
which are available as intra-vaginal creams or pessaries, or
are taken orally.
23
The anticandidal activity demonstrated by the products
tested in this study indicates that formulated tea tree oil
appears to retain the activity of non-formulated tea tree
oil. Demonstrating that products were active in vitro
suggests that these products are potentially useful in vivo.
However, in-vivo effectiveness can only be determined
with a randomized clinical trial to evaluate each product
thoroughly. I t is also possible that product ingredients
other than the tea tree oil, may have anti-candida activity,
and are contributing to the total activity of the product.
However, this is unlikely as the tea tree oil MI Cs were the
same for pure tea tree oil and for product. To determine
accurately any synergic or antagonistic effect of product
excipients it will be necessary to perform chequerboard
analyses.
I deally, a vaginal candidiasis treatment would effect a
mycological cure as well as relieve the candidiasis-
associated symptoms of vulval and vaginal pruritis, vulval
erythema and discharge. There are isolated case reports
suggesting that tea tree oil products may assist in relieving
the symptoms of pruritis and vaginal burning.
13,24
These
preliminary claims are promising and future investigations
may provide support for these observations.
One issue that may require further investigation is that
of adverse reactions to tea tree oil. Pea
22
investigated the
use of intravaginal tea tree oil preparations for a variety of
vaginal conditions and found no vaginal irritation in 130
trial participants. Similarly, a study by Barnes
13
found no
vaginal irritation in 20 trial participants. The incidence of
adverse reactions to tea tree oil in the general population
is as yet unknown, but these studies suggest that the
incidence of vaginal irritation may be relatively low.
I n summary, tea tree oil shows signicant promise as a
potential therapeutic agent for the treatment of vaginal
candidiasis. I n-vitro results indicate susceptibility at low
concentrations ( 0.5% v/v) and products containing tea
tree oil maintained in-vitro efcacy. What is now required
is clinical trials to determine the usefulness of tea tree oil
in vivo.
Acknowledgements
The assistance of the Division of Microbiology and
I nfectious Diseases, The Western Australian Centre for
Pathology and Medical Research, Nedlands, in obtaining
clinical isolates is appreciated. This work was supported by
Australian Bodycare Corporation Pty. Ltd, Currumbin,
QLD, Australia and, in part, by a grant from the Rural
I ndustries Research and Development Corporation
(UWA-24A).
References
1. Stanway, A. (1982). Alternative Medicine: A Guide to Natural
Therapies. Penguin Books Ltd, Harmondsworth.
2. Hoffman, D. L. (1987). The Herb Users Guide. Thorsons
Publishing Group, Wellingborough, Northamptonshire.
3. Tyler, V. E. (1993). The Honest Herbal, 3rd edn. Haworth Press
Inc., Binghampton, NY.
4. Marti, J. (1995). The Alternative Health and Medicine
Encyclopedia. Gale Research International Inc., Detroit, MI.
5. Altman, P. M. (1988). Australian tea tree oil. Australian Journal
of Pharmacy 69, 2768.
6. Carson, C. F. & Riley, T. V. (1994). Susceptibility of Propioni-
bacterium acnes to the essential oil of Melaleuca alternifolia.
Letters in Applied Microbiology 19, 245.
7. Carson, C. F., Cookson, B. D., Farrelly, H. D. & Riley, T. V.
(1995). Susceptibility of methicillin-resistant Staphylococcus aureus
to the essential oil of Melaleuca alternifolia. Journal of Antimicrobial
Chemotherapy 35, 4214.
8. Hammer, K. A., Carson, C. F. & Riley, T. V. (1996).
Susceptibility of transient and commensal skin ora to the essential
oil of Melaleuca alternifolia (tea tree oil). American Journal of
Infection Control 24, 1869.
594
Activity of essential oils against Candida spp. in vitro
9. Nyirjesy, P., Seeney, S. M., Grody, M. H. T., Jordan, C. A. &
Buckley, H. R. (1995). Chronic fungal vaginitis: the value of
cultures. American Journal of Obstetrics and Gynecology 173,
8203.
10. Mrdh, P.-A. (1991). The vaginal ecosystem. American
Journal of Obstetrics and Gynecology165, 11638.
11. Horowitz, B. J., Giaquinta, D. & Ito, S. (1992). Evolving patho-
gens in vulvovaginal candidiasis: implications for patient care.
Journal of Clinical Pharmacology 32, 24855.
12. Newall, C. A., Anderson, L. A. & Phillipson, J. D. (1996).
Herbal Medicines. A Guide for Health-care Professionals. The
Pharmaceutical Press, London.
13. Barnes, B. (1989). The development of topical applications
containing tea tree oil for vaginal conditions. In Modern
PhytotherapyThe Clinical Signicance of Tea Tree Oil and Other
Essential Oils. Proceedings of a Conference in Sydney, September
17 1989, Vol. I, pp. 2735. Australian Tea Tree Industry Associa-
tion, Coraki, Australia.
14. Blackwell, A. L. (1991). Tea tree oil and anaerobic (bacterial)
vaginosis. Lancet 337, 300.
15. Crissey, J. T., Lang, H. & Parish, L. C. (1995). Manual of
Medical Mycology. Blackwell Science, Cambridge, MA.
16. Beylier, M. F. (1979). Bacteriostatic activity of some Australian
essential oils. Perfumer and Flavorist 4, 235.
17. Morris, J. A., Khettry, A. & Seitz, E. W. (1979). Antimicrobial
activity of aroma chemicals and essential oils. Journal of the
American Oil Chemists Society56, 595603.
18. Yousef, R. T. & Tawil, G. G. (1980). Antimicrobial activity of
volatile oils. Pharmazie 35, 698701.
19. Hili, P., Evans, C. S. & Veness, R. G. (1997). Antimicrobial
action of essential oils: the effect of dimethylsulphoxide on the
activity of cinnamon oil. Letters in Applied Microbiology 24, 26975.
20. Southwell, I. A., Hayes, A. J., Markham, J. & Leach, D. N.
(1993). The search for optimally bioactive Australian tea tree oil.
Acta Horticulturae 344, 25665.
21. Nenoff, P., Haustein, U. F. & Brandt, W. (1996). Antifungal
activity of the essential oil of Melaleuca alternifolia (tea tree oil)
against pathogenic fungi in vitro. Skin Pharmacology 9, 38894.
22. Pea, E. F. (1962). Melaleuca alternifolia oil. Its use for
trichomonal vaginitis and other vaginal infections. Obstetrics and
Gynecology 19, 7935.
23. Working Group of the British Society for Medical Mycology.
(1995). Management of genital candidiasis. British Medical Journal
310, 12414.
24. Merkur, H. (1989). The impact of HPV (human papilloma virus)
in gynaecology. In Modern Phytotherapy The Clinical Signicance
of Tea Tree Oil and Other Essential Oils. Proceedings of a Con-
ference in Sydney, September 17 1989. Vol. I, pp. 3641. Aus-
tralian Tea Tree Industry Association, Coraki, Australia.
Received 9 February 1998; accepted 27 May 1998
595